“Everything starts out looking like a toy” (No. 10)
Observations at the intersection of data and systems.
This week’s toy: a keychain version of Super Mario 2. There’s only one problem with this: since we’re not leaving the house, we don’t need keys. This is edition No. 10, coming at you on September 5, 2020.
The Big Idea
Data operations describes people, process, and tools that enable data to move seamlessly. But we are really lousy at diagramming and documenting that system such that we get a consistent output when we add new input. Maxime Beauchamin writes in a fantastic and dense article about data engineering,
To put it simply, immutable data along with versioned logic are key to reproducibility.
This works really well within a controlled environment when you know what data to expect. But what about in a typical go-to-market sales environment, when you might get input from almost any system (with almost any value). In that situation you need both the static definition of the system and the dynamic definition of the transformation that gets applied to data as it enters and transits your system.
A person asking “what happens when a lead fills out a landing page form and doesn’t capitalize their name” should know whether or not the expected behavior includes 100% capitalization on names in a CRM. That person also needs to know if and when data errors prevent that information from being viable until enriched.
What’s the takeaway? The underlying data in a system shouldn’t change just because we want to analyze it differently. One challenge that organizations have is mixing the data with transformation, instead of clearly stating the definition of the data and what’s being done to produce the result.
We’d like to know …
Mobile phone purchase is a decent way to gauge discretionary spending. Are you buying a new one this year?
Click the tweet to vote.
Links for Reading and Sharing
These are links that caught my eye.
1/ The World of the App Store - have you ever wondered about the developer arrangement Apple has with the developers when they host apps on the App Store?
Some of the app behavior you’ve seen is related to this relationship. Tl; dr: when the publisher makes money off of a mobile app, the platform wants a cut.
2/ Try on clothing without going to the store - this project uses a single reference image to remap clothing on a person. Put another way, your favorite retailer could ask you to send them a photo in your favorite outfit and suggest a new one for you while showing you how you look it it!
3/ Virtua Dog - Ever wanted your pet in a video game? It might be possible soon for Rover to join you in Animal Crossing. The trick might be convincing your dog to wear the motion capture suit.
On the Reading List
The Data Warehouse Toolkit - Ralph Kimball and Margy Ross provide clear guidelines for building the best way to store your data.
Storytelling with Data: Let’s Practice! Cole Nussbaumer Knaflic shares scenarios for data visualization and how to make your visuals better.
What to do next
Hit reply if you’ve got links to share, data stories, or want to say hello.
I’m grateful you read this far. Thank you.
If you found this useful, consider sharing with a friend.
Want essays? Check out on Data Operations or other writings at gregmeyer.com.
The next big thing always starts out being dismissed as a “toy.” - Chris Dixon